AI/ML, distributed computing, and blockchained based reservoir management platform

Abstract

A system, for controlling well site operations, comprising a machine learning engine, a predictive engine, a node system stack, and a blockchain. The learning engine includes a machine learning algorithm, an algorithmically generated earth model, and control variables. The learning algorithm generates a trained data model using the algorithmically generated earth model. The predictive engine includes an Artificial Intelligence (AI) algorithm. The AI algorithm generates a trained AI algorithm using the trained data model and earth model variables using the trained AI algorithm. The system stack is communicable coupled to the predictive engine, the learning engine, the blockchain, sensors, and a machine controller. The blockchain having a genesis block and a plurality of subsequent blocks. Each subsequent block comprising a well site entry and a hash of a previous entry. The well site entry comprises transacted operation control variables. The transacted variables are based on the generated earth model variables.

What is claimed is: 
     
       1. A system stack for managing well site operations, the system stack comprising:
 at least one node system stack; and 
 at least one predictive engine that includes a drill path and production control pattern recognition component and at least one machine learning engine,
 wherein the at least one machine learning engine has at least one machine learning algorithm, at least one algorithmically generated earth model, and receives at least one operation control variable from a machine controller configured to control equipment for well site operations, wherein the at least one machine learning engine is configured to generate at least one trained data model, 
 wherein the at least one predictive engine has an artificial intelligence algorithm, 
 wherein the at least one predictive engine is configured to generate a trained artificial intelligence algorithm using the artificial intelligence algorithm and the at least one trained data model generated by the at least one machine learning engine, 
 wherein the trained artificial intelligence algorithm has a parameter set that is less than a parameter set used by the artificial intelligence algorithm, 
 wherein the trained artificial intelligence algorithm is configured to generate earth model variables, 
 wherein the at least one node system stack is coupled to the at least one predictive engine, the at least one machine learning algorithm, a distributed network, a plurality of sensors, and the at least one machine controller for communication therewith, 
 wherein the distributed network includes a genesis block chained to a plurality of subsequent blocks, 
 wherein each of the plurality of subsequent blocks includes a well site entry and a cryptographic hash value of a previous well site entry, 
 wherein the well site entry includes at least one operation control variable and a well site operation from the machine controller configured to control equipment for well site operations, wherein the at least one operation control variable stored in the well site entry is, at least in part, based on at least one of the generated earth model variables; 
 
 a visualization engine configured to generate a display of a drill path, the received at least one operation control variable, and the generated earth model variables; and 
 an optimization engine configured to optimize the generated earth model variables by sampling the generated earth model variables based on at least one drilling model and an optimization tool configured to predict at least one optimized drill path,
 wherein the predicted at least one optimized drill path is based on one or more objective criteria including a shortest length, minimum drilling time, maximum Rate Of Penetration, minimum bit wear, minimum mud loss, minimum overall drilling cost, minimum curvature, complexity of the drill path, and maximum safety, and 
 wherein the visualization engine updates the display based on the predicted at least one optimized drill path. 
 
 
     
     
       2. The system stack of  claim 1 , wherein the system stack is executed on a hardware node, partitions of a hardware node, a plurality of hardware nodes, or a combination thereof. 
     
     
       3. The system stack of  claim 2 , wherein the system stack includes a plurality of partitions and the stack comprises a middleware controller coupled to each partition for communication with components included in the plurality of partitions including node system stacks, predictive engines, and machine learning engines. 
     
     
       4. The system stack of  claim 3 , wherein the middleware controller is a Robot Operating System (ROS) based controller. 
     
     
       5. The system stack of  claim 1 , wherein the optimization engine performs one of a Bayesian optimization, genetic algorithm optimization, and particle swarm optimization. 
     
     
       6. The system stack of  claim 1 , further comprising:
 a deep particle filter configured to clean the well log data variables and seismic data variables received from sensors and well site operation equipment; and 
 a forward modeling component to compare predicted variables in the generated earth model to the cleaned well log data variables and seismic data variables. 
 
     
     
       7. An apparatus for managing well site operations, the apparatus comprising:
 at least one node system stack; and 
 at least one predictive engine that includes a drill path and production control pattern recognition component and at least one machine learning engine,
 wherein the at least one machine learning engine has at least one machine learning algorithm, at least one algorithmically generated earth model, and at least one operation control variable from a machine controller configured to control equipment for well site operations, wherein the at least one machine learning engine is configured to generate at least one trained data model, 
 wherein the at least one predictive engine has an artificial intelligence algorithm, 
 wherein the at least one predictive engine is configured to generate a trained artificial intelligence algorithm using the artificial intelligence algorithm and the at least one trained data model generated by the at least one machine learning engine, 
 wherein the trained artificial intelligence algorithm has a parameter set that is less than a parameter set used by the artificial intelligence algorithm, 
 wherein the trained artificial intelligence algorithm is configured to generate earth model variables, 
 wherein the at least one node system stack is coupled to the at least one predictive engine, the at least one machine learning algorithm, a distributed network, a plurality of sensors, and the at least one machine controller for communication therewith; 
 
 a visualization engine configured to generate a display of a drill path, the received at least one operation control variable, and the generated earth model variables; 
 an optimization engine configured to optimize the generated earth model variables by sampling the generated earth model variables based on at least one drilling model and an optimization tool configured to predict at least one optimized drill path,
 wherein the predicted at least one optimized drill path is based on one or more objective criteria including a shortest length, minimum drilling time, maximum Rate Of Penetration, minimum bit wear, minimum mud loss, minimum overall drilling cost, minimum curvature, complexity of the drill path, and maximum safety, and 
 wherein the visualization engine updates the display based on the predicted at least one optimized drill path. 
 
 
     
     
       8. The apparatus of  claim 7 , wherein the apparatus is executed on a hardware node, partitions of a hardware node, a plurality of hardware nodes, or a combination thereof. 
     
     
       9. The apparatus of  claim 8 , wherein the apparatus includes a plurality of partitions and the stack further comprises a middleware controller coupled to each partition for communication with components included in the plurality of partitions including node system stacks, predictive engines, and machine learning engines. 
     
     
       10. The apparatus of  claim 9 , wherein the middleware controller is a Robot Operating System (ROS) based controller. 
     
     
       11. The apparatus of  claim 7 , wherein the optimization engine performs one of a Bayesian optimization, genetic algorithm optimization, and particle swarm optimization. 
     
     
       12. The apparatus of  claim 7 , further comprising:
 a deep particle filter configured to clean the well log data variables and seismic data variables received from sensors and well site operation equipment; and 
 a forward modeling component to compare predicted variables in the generated earth model to the cleaned well log data variables and seismic data variables. 
 
     
     
       13. A method for managing well site operations using at least one system stack that includes at least one node system stack, the method comprising:
 receiving, by at least one machine learning engine, at least one operation control variable from a machine controller configured to control equipment for well site operations, wherein the at least one machine learning engine is configured to generate at least one trained data model, wherein the at least one machine learning is part of at least one predictive engine that also includes a drill path and production control pattern recognition component, wherein the at least one machine learning engine has at least one machine learning algorithm, at least one algorithmically generated earth model; 
 generating a trained artificial intelligence algorithm using an artificial intelligence algorithm of the at least one predictive engine has an artificial intelligence algorithm and the at least one trained data model generated by the at least one machine learning engine, wherein the trained artificial intelligence algorithm has a parameter set that is less than a parameter set used by the artificial intelligence algorithm, wherein the trained artificial intelligence algorithm is configured to generate earth model variables,
 wherein the at least one node system stack is coupled to the at least one predictive engine, the at least one machine learning algorithm, a distributed network, a plurality of sensors, and the at least one machine controller for communication therewith, wherein the distributed network includes a genesis block chained to a plurality of subsequent blocks, 
 wherein each of the plurality of subsequent blocks includes a well site entry and a cryptographic hash value of a previous well site entry, wherein the well site entry includes at least one operation control variable and a well site operation from the machine controller configured to control equipment for well site operations, wherein the at least one operation control variable stored in the well site entry is, at least in part, based on at least one of the generated earth model variables; 
 
 generating, by a visualization engine, a display of a drill path, the received at least one operation control variable, and the generated earth model variables; 
 optimizing, by an optimization engine, the generated earth model variables by sampling the generated earth model variables based on at least one drilling model and an optimization tool configured to predict at least one optimized drill path,
 wherein the predicted at least one optimized drill path is based on one or more objective criteria including a shortest length, minimum drilling time, maximum Rate Of Penetration, minimum bit wear, minimum mud loss, minimum overall drilling cost, minimum curvature, complexity of the drill path, and maximum safety; 
 controlling at least one well site operation using the at least one operation control variable included in the well site entry; and 
 updating, by the visualization engine, the display based on the predicted at least one optimized drill path. 
 
 
     
     
       14. The method of  claim 13 , wherein the system stack is executed on a hardware node, partitions of a hardware node, a plurality of hardware nodes, or a combination thereof. 
     
     
       15. The method of  claim 14 , wherein the system stack includes a plurality of partitions and the stack further comprises a middleware controller coupled to each partition for communication with components included in the plurality of partitions including system stacks, predictive engines, and machine learning engines. 
     
     
       16. The method of  claim 15 , wherein the middleware controller is a Robot Operating System (ROS) based controller. 
     
     
       17. The method of  claim 13 , wherein the optimization tool performs one of a Bayesian optimization, genetic algorithm optimization, and particle swarm optimization. 
     
     
       18. The method of  claim 13 , further comprising cleaning the well log data variables and seismic data variables received from sensors and well site operation equipment using a deep particle filter; and comparing predicted variables in the generated earth model to the cleaned well log data variables and seismic data variables using a forward modeling component.